Amazon Jobs Jobs API.

import requests

BASE = "https://www.amazon.jobs"

def fetch_page(offset: int, limit: int = 100, country: str = "GBR") -> dict:
    resp = requests.get(
        f"{BASE}/en/search.json",
        params={"offset": offset, "result_limit": limit, "country": country},
        headers={"Accept": "application/json"},
        timeout=30,
    )
    resp.raise_for_status()
    return resp.json()

data = fetch_page(0)
print(data["hits"], "total jobs")

import time

def iter_jobs(country: str = "GBR", limit: int = 100):
    offset = 0
    while True:
        data = fetch_page(offset, limit, country)
        jobs = data.get("jobs", [])
        if not jobs:
            break
        for job in jobs:
            yield job
        offset += len(jobs)
        if offset >= data.get("hits", 0):
            break
        time.sleep(0.1)  # keep a single, throttled request in flight

import re
from urllib.parse import urljoin

ID_RE = re.compile(r"/jobs/(?P<id>[A-Za-z0-9]+)")

def parse_core(job: dict) -> dict:
    job_path = job.get("job_path") or ""
    listing_url = urljoin("https://www.amazon.jobs", job_path)
    match = ID_RE.search(job_path)
    path_id = match.group("id") if match else None
    external_id = (job.get("id_icims") or path_id or "").strip() or None
    return {
        "external_id": external_id,
        "listing_url": listing_url,
        "title": (job.get("title") or "").strip() or None,
        "company_name": job.get("company_name"),
        "posted_at": job.get("posted_date"),
        "job_category": job.get("job_category"),
        "schedule_type": job.get("job_schedule_type"),
    }

import json

def parse_locations(job: dict) -> list[dict]:
    out = []
    for raw in job.get("locations") or []:
        raw = (raw or "").strip()
        if raw.startswith("{") or raw.startswith("["):
            try:
                payload = json.loads(raw)
            except json.JSONDecodeError:
                continue
            city = payload.get("normalizedCityName") or payload.get("city")
            state = payload.get("normalizedStateName")
            country = payload.get("normalizedCountryName")
            text = ", ".join(p for p in (city, state, country) if p)
            is_remote = (payload.get("type") or "").upper() in ("REMOTE", "VIRTUAL")
            out.append({"text": text, "city": city, "state": state,
                        "country": country, "is_remote": is_remote})
        elif raw:
            out.append({"text": raw})
    if not out and job.get("location"):
        out.append({"text": job["location"]})
    return out

def build_description(job: dict) -> str:
    parts = [job.get("description") or ""]
    if job.get("basic_qualifications"):
        parts.append("<h2>Basic qualifications</h2>" + job["basic_qualifications"])
    if job.get("preferred_qualifications"):
        parts.append("<h2>Preferred qualifications</h2>" + job["preferred_qualifications"])
    return "<br/><br/>".join(p for p in parts if p)

for job in iter_jobs("GBR"):
    record = parse_core(job)
    record["locations"] = parse_locations(job)
    record["description_html"] = build_description(job)
    print(record["external_id"], record["title"])